Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
DuckDB CLI runs SQL queries against CSV, Parquet, and JSON files directly from the terminal without a database server. It also converts between file formats and integrates with shell pipes and scripts.
Queries files in place without loading them into a running database, so exploratory analysis is faster than spinning up a server or writing a pandas script.
When to use it
- Querying a CSV file for records matching specific criteria without opening a spreadsheet
- Converting Parquet exports to CSV for use in Excel or Google Sheets
- Aggregating transaction totals by category across multiple log files
- Joining orders and customer data stored in different file formats
- Previewing a large dataset with a quick LIMIT query before a full import
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Show me total sales per category from sales_2024.csv, sorted highest to lowest
- 1Identify the target file and the required grouping and aggregation
- 2Construct a SELECT with GROUP BY category and SUM(amount) ORDER BY total DESC
- 3Run: duckdb -table -c "SELECT category, SUM(amount) AS total FROM 'sales_2024.csv' GROUP BY category ORDER BY total DESC"
- 4Return the formatted ASCII table output
A table listing each category with its total sales, ordered from highest to lowest
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.